Investigators in Project 3 (M. Brenner and B. Savoldo) have tested the safety and antitumor efficacy of T cells expressing a chimeric antigen receptor (CAR) specific for the neuroblastoma (NB)-associated antigen GD2, demonstrating tumor responses in 5/11 patients with refractory/relapsed NB and complete remissions in 3. This study also indicated the importance of costimulatory signals provided to T cells through engagement of their native receptors by specific antigens (in this case by EBV-derived peptides) on host antigen-presenting cells. To reproduce and enhance these effects while avoiding the concomitant complexity of manufacturing virus-specific CTLs and the difficulty of treating EBV-seronegative patients, the investigators predict that engineering the CAR.GD2 molecule to incorporate two costimulatory endodomains in order to enhance CAR+ T-cell expansion and persistence in vivo will improve antitumor efficacy and clinical outcome. Because of the potential toxicities associated with excessive or accelerated T-cell activities, they also incorporated an inducible (dimerizable) caspase-9 (iC9) protein, which rapidly ablated T-cell activity and associated graft-versus -host disease in an earlier trial. Aim 1 will test this approach in patients with relapsed/refractory NB, using T cells expressing a CAR.GD2 that incorporates the CD28 and OX40 costimulatory endodomains and the iC9 gene suicide switch. The expansion, long-term persistence, and antitumor effects of these engineered T cells will be assessed in Aim 2. Finally, in Aim 3, a mouse model will be used to ask if simultaneously targeting, through native and chimeric receptors, two antigens independently expressed by NB cells will prevent tumor immune escape arising from single antigen loss and will also promote tumor-associated antigen presentation by upregulation of HLA molecules due to the inflammatory cytokines locally secreted by CAR+ T cells. Specifically, we will explore the potential for synergy between GD2.CAR+ T cells and T cells that are transgenic for a high-affinity native (aP) TCR specific for an HI_A-A2-restricted epitope on the cancer testis antigen PRAME, which is abundantly expressed by most NB cells. In our previous funding period project we made use of the EBV-specific CTLs (expressing CAR-GD2) developed by Project 1. In the current proposal we use the iC9 safety gene developed by Project 1 and provide our (3D2-CAR to project 1 in return. If successful, the concept of directing both native and chimeric T cell receptor to tumor cell targets may be adapted to other tumors.